Inspiration
I didn't have any inspiration. The reason why I chose use case 2 was because everyone was choosing use case 1 and I wanted to be different
What it does
It gathers all USA zip codes with a median annual income of 30,000$ or less and then groups all neighboring zip codes. It passes it as the input to the AI and the AI provides the best locations to place First Tee branches so that underserved communities can have easy access.
How we built it
I used the library USZipCodes to find all zip codes with a median annual salary of 30,000$ and less. I built a game for the AI to play where it got points for locations touching parts of zip codes within certain distances but lost points if the building reaches over lapped.
Challenges we ran into
Due to the lack of in-depth data and computer power, I made a rudimentary AI that looks at positional data of zipcodes but none of the elevation, rivers, streets, etc. I also had to crop all maps to a 5x5 grid due to the lack of computer power and time.
Accomplishments that we're proud of
A working AI that if trained more with more data, would only become stronger.
What we learned
I learned how to leverage Zip code data to my advantage and how to use a new library that I found to use the zip code data.
What's next for Use Case 2 Hunter Ford Submission
Add more data such as elevation maps, rivers, streets, and where we can build, while also not cropping the maps.
Log in or sign up for Devpost to join the conversation.